This project reflects a longstanding NLM interest in clinical terminology and message standards. It uses and tunes the message and vocabulary standards that NLM has supported to facilitate interoperability, defined as communicating data from a source to a destination where it can be used in computing. (HIMSS Dictionary of Healthcare Information Technology Terms, Acronyms and Organizations, 2nd Edition, 2010, Appendix B, p190.) NLM has been at the nexus of standards needed to make clinical data flow from sources for clinical care and epidemiologic databases. The major code systems required by Meaningful Use regulations, LOINC, SNOMED, RxNorm, and UCUM, are all developed and/or supported by NLM. We are directly engaged in work on HL7 electronic messaging standards related to Clinical Genomics Structured Reporting, and reporting results of newborn screening (recently published in the HL7 V2 genetic reporting implementation guide as part of the Laboratory Results Interface (LRI) of HL7). NLM is developing standards-based tools and techniques that can be used in EHRs, PHRs, and research. We have made these tools open source and freely-available by publishing the source code on GitHub. They are Section 508-compliant (i.e., accessible to screen readers). Groups or institutions can customize many of the features for their particular needs. A) LHC-Forms: Form Rendering Widget LHC-Forms creates input forms for Web-based medical applications, EHRs, PHRs, and mobile health apps. LHC-Forms can render a powerful data entry form for laboratory panels, survey instruments, etc. from any of the 2,000+ panels defined in LOINC. Implementers can also use it to develop forms based on their own arbitrary variables. The Office of Management and Budget (OMB) has adopted LHC-Forms for a pilot testing its use in its large data systems. Programmers interested in using LHC-Forms in their own EMR, PHR, or other application can download the LHC-Forms software and view the documentation at http://lhncbc.github.io/lforms/. LHC-Forms supports many form attributes, including: data type, cardinality, default value, units of measure (if numeric), answer lists and multiple choice/multiple answer variables, relationship (in a nested hierarchy) to other questions, scoring of survey instruments, default value settings, validation checks to ensure quality data collection, skip logic and help messages. LHC-Forms uses the NLM-developed autocompleter package (http://lhncbc.github.io/autocomplete-lhc/). LHC-Forms has the ability to accept, store, and display FHIR Questionnaire resources (SDC and standard) as well as output FHIR DiagnosticReport and QuestionnaireResponse messages and convert to an HL7 v2 message. We lead a new HL7 FHIR work group (Structured Data Capture (SDC)) to standardize questionnaires and associated tools. This specification is being balloted under an ANSI approved process at the present time and anticipate having a first version in January 2019. Work to support FHIRPath for defining Questionnaire features like score rules and auto-population of data has begun, starting with collaboration on a new open source version of JavaScript FHIRPath library, https://github.com/lhncbc/fhirpath.js. Our related Form Builder tool enables users to build and customize forms using an informative dashboard with real-time updates that display user selections. It can also output the built form definitions as FHIR Questionnaires, and store and retrieve Questionnaires from a selected FHIR server. Try it: https://lhcformbuilder.nlm.nih.gov A new widget, complementary to the LHC-Forms widget, is being developed to show flowsheets of patient data, which are tables where various test and measure names are on the left and the columns are the dates on which the measures were taken. The new widget will be used by a demo application that will pull data from a FHIR server. B) Clinical Table Search Service The Clinical Table Search Service provides look up functions to tables such as master files and coding systems, in order to provide answer lists for fields in many forms. The service connects tables to fields in the form by URLs whose parameters control what table to search, which table fields to return to the choice menu grid, and which fields of the selected item to store as hidden content in the form input fields. We now have over 20 data tables that users can browse and use in applications. The coding systems include most of the major genetic databases from NCBI as well as from Cosmic Cancer Mutation and EBI. Our implementation provides API access to many clinical tables: LOINC, RxTerms, ICD-10-CM, most NCBI genomics tables, COSMIC, National Provider Identifiers, and others, with data now updated monthly. Try it: https://clinicaltables.nlm.nih.gov. C) Validator and converter for UCUM units of measure. Standard units of measure are essential to the use of numeric variables in clinical care and research. Currently, they are not standardized and use varying strings. In a large sample of HL7 messages, we found 60 different ways to express the units for a red blood cell count. The Universal Code for Reporting Units of Measure (UCUM, http://unitsofmeasure.org/) was developed to solve this problem. Standard units are needed to exchange and aggregate numeric values (e.g. laboratory test results and vital signs), or to utilize them for clinical decision support. UCUM has been adopted by most clinical standards organizations, including ANSI approved standards development orgs (SDOs): HL7, IEEE (instrument measurements), DICOM (radiology measurement), and ISO-11240 for development of medicinal products. It is required by meaningful use for HL7 for Public Health laboratory reporting, and for most measurements in HL7s CDA reporting. We developed an open source JavaScript UCUM validator and conversion library that will verify that units strings claiming to be UCUM units are valid UCUM units, can batch validate the units in a table submitted as a CSV file, and can also convert values reported as one specific UCUM unit to another commensurate UCUM unit (e.g. ounces to kilograms), which could help aggregate or analyze data from multiple sources. Try it: https://ucum.nlm.nih.gov/ucum-lhc/demo.html. The library now has a new suggestions feature to help users find UCUM codes, and can now be downloaded and installed via the npm and bower package managers. Additionally, we have added a web API service for UCUM validation and conversion, based on code donated by Jozef Aerts of xml4pharma.com. A new website was created as an entry point for all of our UCUM resources (the web service API, the library, and the Clinical Table Search Service API). Try it: https://ucum.nlm.nih.gov D) Research database of deidentified FHIR resources A set of deceased, deidentified patient data for 10,000 patients was obtained from Regenstrief Institute. These were then loaded into a HAPI FHIR server to test its performance. Significant performance issues were encountered both with loading the data and searching, and various attempts were made to solve the issue including switching databases. In FY2018, we began working with contractors from Simpatico, who are revising the HAPI FHIR code to remove performance bottlenecks.